orca-sdk 0.1.2__py3-none-any.whl → 0.1.4__py3-none-any.whl

orca_sdk/client.py

@@ -6,15 +6,31 @@ import json
 import logging
 import os
 import uuid
+from contextlib import contextmanager
+from contextvars import ContextVar
 from string import Formatter
-from typing import Any, Literal, Mapping, NotRequired, TypedDict, cast, overload
+from typing import (
+    Any,
+    Callable,
+    Generator,
+    Literal,
+    Mapping,
+    NotRequired,
+    Self,
+    TypedDict,
+    cast,
+    overload,
+)
 
 from httpx import (
     URL,
     USE_CLIENT_DEFAULT,
+    BaseTransport,
     Client,
     Headers,
     HTTPTransport,
+    Limits,
+    Proxy,
     Request,
     Response,
     Timeout,
@@ -117,6 +133,8 @@ class ClassificationEvaluationRequest(TypedDict):
     datasource_value_column: str
     record_telemetry: NotRequired[bool]
     telemetry_tags: NotRequired[list[str] | None]
+    subsample: NotRequired[int | float | None]
+    ignore_unlabeled: NotRequired[bool]
 
 
 class CleanupResponse(TypedDict):
@@ -145,6 +163,7 @@ class CountPredictionsRequest(TypedDict):
     prediction_ids: NotRequired[list[str] | None]
     start_timestamp: NotRequired[str | None]
     end_timestamp: NotRequired[str | None]
+    memory_id: NotRequired[str | None]
 
 
 class CreateApiKeyRequest(TypedDict):
@@ -175,6 +194,12 @@ class CreateOrgPlanRequest(TypedDict):
     tier: Literal["FREE", "PRO", "ENTERPRISE", "CANCELLED"]
 
 
+class DatasetFilterItem(TypedDict):
+    field: str
+    op: Literal["==", "!=", ">", ">=", "<", "<=", "in", "not in", "like"]
+    value: Any
+
+
 class DeleteMemoriesRequest(TypedDict):
     memory_ids: list[str]
 
@@ -192,7 +217,7 @@ class EmbedRequest(TypedDict):
 class EmbeddingEvaluationRequest(TypedDict):
     datasource_name_or_id: str
     eval_datasource_name_or_id: NotRequired[str | None]
-    subsample: NotRequired[int | None]
+    subsample: NotRequired[int | float | None]
     datasource_value_column: NotRequired[str]
     datasource_label_column: NotRequired[str | None]
     datasource_score_column: NotRequired[str | None]
@@ -201,7 +226,7 @@ class EmbeddingEvaluationRequest(TypedDict):
     weigh_memories: NotRequired[bool]
 
 
-EmbeddingFinetuningMethod = Literal["classification", "batch_triplet_loss"]
+EmbeddingFinetuningMethod = Literal["classification", "regression", "batch_triplet_loss"]
 
 
 class FeedbackMetrics(TypedDict):
@@ -215,7 +240,19 @@ FeedbackType = Literal["CONTINUOUS", "BINARY"]
 class FilterItem(TypedDict):
     field: list
     op: Literal["==", "!=", ">", ">=", "<", "<=", "in", "not in", "like"]
-    value: str | int | float | bool | list[str] | list[int] | list[float] | list[bool] | None
+    value: str | int | float | bool | list[str | None] | list[int] | list[float] | list[bool] | None
+
+
+class GetDatasourceRowCountRequest(TypedDict):
+    filters: NotRequired[list[DatasetFilterItem]]
+
+
+class GetDatasourceRowsRequest(TypedDict):
+    filters: NotRequired[list[DatasetFilterItem]]
+    limit: NotRequired[int]
+    offset: NotRequired[int]
+    shuffle: NotRequired[bool]
+    shuffle_seed: NotRequired[int | None]
 
 
 class GetMemoriesRequest(TypedDict):
@@ -236,6 +273,18 @@ class InternalServerErrorResponse(TypedDict):
     request_id: str
 
 
+JobStatus = Literal["INITIALIZED", "DISPATCHED", "WAITING", "PROCESSING", "COMPLETED", "FAILED", "ABORTING", "ABORTED"]
+
+
+class JobStatusInfo(TypedDict):
+    status: JobStatus
+    steps_total: int | None
+    steps_completed: int | None
+    exception: str | None
+    updated_at: str
+    created_at: str
+
+
 class LabelClassMetrics(TypedDict):
     label: int | None
     label_name: NotRequired[str | None]
@@ -243,6 +292,11 @@ class LabelClassMetrics(TypedDict):
     memory_count: int
 
 
+class LabelPercentage(TypedDict):
+    label: int | None
+    percentage: float
+
+
 class LabeledExample(TypedDict):
     text: str
     label_name: str
@@ -253,6 +307,7 @@ class LabeledMemoryInsert(TypedDict):
     value: str | bytes
     metadata: NotRequired[dict[str, str | int | float | bool | None]]
     source_id: NotRequired[str | None]
+    partition_id: NotRequired[str | None]
     label: int | None
 
 
@@ -317,8 +372,6 @@ class MemorysetClassPatternsMetrics(TypedDict):
 class MemorysetClusterAnalysisConfig(TypedDict):
     min_cluster_size: NotRequired[int | None]
     max_cluster_size: NotRequired[int | None]
-    clustering_method: NotRequired[Literal["density", "graph"]]
-    min_cluster_distance: NotRequired[float]
     partitioning_method: NotRequired[Literal["ng", "rb", "cpm"]]
     resolution: NotRequired[float | None]
     num_iterations: NotRequired[int]
@@ -335,16 +388,32 @@ class MemorysetClusterMetrics(TypedDict):
 class MemorysetConceptAnalysisConfig(TypedDict):
     high_level_description: NotRequired[str | None]
     max_sample_rows: NotRequired[int]
+    min_lookup_count: NotRequired[int]
+    max_lookup_count: NotRequired[int]
+    overlap_decay_tau: NotRequired[float]
+    target_lcc_fraction: NotRequired[float]
+    min_cluster_count: NotRequired[int]
+    max_cluster_count: NotRequired[int]
     max_trial_count: NotRequired[int]
-    min_desired_clusters_per_label: NotRequired[int]
-    max_desired_clusters_per_label: NotRequired[int]
-    accuracy_importance: NotRequired[float]
-    noise_penalty: NotRequired[float]
+    resolution: NotRequired[float]
+    iterations: NotRequired[int]
+    use_generative_naming: NotRequired[bool]
     naming_examples_count: NotRequired[int]
     naming_counterexample_count: NotRequired[int]
+    primary_label_pct_threshold: NotRequired[float]
     seed: NotRequired[int]
 
 
+class MemorysetDistributionAnalysisConfig(TypedDict):
+    neighbor_counts: NotRequired[list[int]]
+    quantiles: NotRequired[list[float]]
+
+
+class MemorysetDistributionMetrics(TypedDict):
+    lookup_score_metrics: dict[str, LookupScoreMetrics]
+    updated_at: str
+
+
 class MemorysetDuplicateAnalysisConfig(TypedDict):
     potential_duplicate_threshold: NotRequired[float]
 
@@ -370,16 +439,6 @@ class MemorysetLabelMetrics(TypedDict):
     updated_at: str
 
 
-class MemorysetNeighborAnalysisConfig(TypedDict):
-    neighbor_counts: NotRequired[list[int]]
-    quantiles: NotRequired[list[float]]
-
-
-class MemorysetNeighborMetrics(TypedDict):
-    lookup_score_metrics: dict[str, LookupScoreMetrics]
-    updated_at: str
-
-
 class MemorysetProjectionAnalysisConfig(TypedDict):
     min_dist: NotRequired[float]
     spread: NotRequired[float]
@@ -411,12 +470,12 @@ class NotFoundErrorResponse(TypedDict):
             "memory",
             "evaluation",
             "analysis",
-            "task",
+            "job",
             "pretrained_embedding_model",
             "finetuned_embedding_model",
             "feedback_category",
-            "embedding_evaluation",
             "org_plan",
+            "worker",
         ]
         | None
     )
@@ -525,6 +584,8 @@ class RegressionEvaluationRequest(TypedDict):
     datasource_value_column: str
     record_telemetry: NotRequired[bool]
     telemetry_tags: NotRequired[list[str] | None]
+    subsample: NotRequired[int | float | None]
+    ignore_unlabeled: NotRequired[bool]
 
 
 class RegressionMetrics(TypedDict):
@@ -567,12 +628,14 @@ class RegressionPredictionRequest(TypedDict):
     prompt: NotRequired[str | None]
     use_lookup_cache: NotRequired[bool]
     consistency_level: NotRequired[Literal["Bounded", "Session", "Strong", "Eventual"] | None]
+    ignore_unlabeled: NotRequired[bool]
 
 
 class ScorePredictionMemoryLookup(TypedDict):
     value: str | bytes
     embedding: list[float]
     source_id: str | None
+    partition_id: str | None
     metadata: dict[str, str | int | float | bool | None]
     memory_id: str
     memory_version: int
@@ -610,6 +673,7 @@ class ScoredMemory(TypedDict):
     value: str | bytes
     embedding: list[float]
     source_id: str | None
+    partition_id: str | None
     metadata: dict[str, str | int | float | bool | None]
     memory_id: str
     memory_version: int
@@ -625,6 +689,7 @@ class ScoredMemoryInsert(TypedDict):
     value: str | bytes
     metadata: NotRequired[dict[str, str | int | float | bool | None]]
     source_id: NotRequired[str | None]
+    partition_id: NotRequired[str | None]
     score: float | None
 
 
@@ -632,6 +697,7 @@ class ScoredMemoryLookup(TypedDict):
     value: str | bytes
     embedding: list[float]
     source_id: str | None
+    partition_id: str | None
     metadata: dict[str, str | int | float | bool | None]
     memory_id: str
     memory_version: int
@@ -648,6 +714,7 @@ class ScoredMemoryUpdate(TypedDict):
     value: NotRequired[str | bytes]
     metadata: NotRequired[dict[str, str | int | float | bool | None] | None]
     source_id: NotRequired[str | None]
+    partition_id: NotRequired[str | None]
     metrics: NotRequired[MemoryMetrics | None]
     score: NotRequired[float | None]
 
@@ -656,6 +723,7 @@ class ScoredMemoryWithFeedbackMetrics(TypedDict):
     value: str | bytes
     embedding: list[float]
     source_id: str | None
+    partition_id: str | None
     metadata: dict[str, str | int | float | bool | None]
     memory_id: str
     memory_version: int
@@ -681,18 +749,6 @@ class SubConceptMetrics(TypedDict):
     memory_count: int
 
 
-TaskStatus = Literal["INITIALIZED", "DISPATCHED", "WAITING", "PROCESSING", "COMPLETED", "FAILED", "ABORTING", "ABORTED"]
-
-
-class TaskStatusInfo(TypedDict):
-    status: TaskStatus
-    steps_total: int | None
-    steps_completed: int | None
-    exception: str | None
-    updated_at: str
-    created_at: str
-
-
 TelemetryField = list
 
 
@@ -733,6 +789,13 @@ class ValidationError(TypedDict):
     type: str
 
 
+WorkerStatus = Literal["IDLE", "BUSY", "DRAINING", "SHUTDOWN", "CRASHED"]
+
+
+class GetTestErrorByStatusCodeParams(TypedDict):
+    status_code: int | Literal["error", "warning"]
+
+
 class DeleteAuthApiKeyByNameOrIdParams(TypedDict):
     name_or_id: str
 
@@ -758,6 +821,10 @@ class DeleteMemorysetByNameOrIdParams(TypedDict):
     name_or_id: str
 
 
+class PostGpuMemorysetByNameOrIdLookupParams(TypedDict):
+    name_or_id: str
+
+
 class GetMemorysetByNameOrIdMemoryByMemoryIdParams(TypedDict):
     name_or_id: str
     memory_id: str
@@ -790,20 +857,35 @@ class PostMemorysetByNameOrIdMemoriesDeleteParams(TypedDict):
     name_or_id: str
 
 
+class PatchGpuMemorysetByNameOrIdMemoryParams(TypedDict):
+    name_or_id: str
+
+
+class PostGpuMemorysetByNameOrIdMemoryParams(TypedDict):
+    name_or_id: str
+
+
+PostGpuMemorysetByNameOrIdMemoryRequest = list[LabeledMemoryInsert] | list[ScoredMemoryInsert]
+
+
+class PatchGpuMemorysetByNameOrIdMemoriesParams(TypedDict):
+    name_or_id: str
+
+
 class PostMemorysetByNameOrIdAnalysisParams(TypedDict):
     name_or_id: str
 
 
 class GetMemorysetByNameOrIdAnalysisParams(TypedDict):
     name_or_id: str
-    status: NotRequired[TaskStatus | None]
+    status: NotRequired[JobStatus | None]
     limit: NotRequired[int | None]
     offset: NotRequired[int | None]
 
 
-class GetMemorysetByNameOrIdAnalysisByAnalysisTaskIdParams(TypedDict):
+class GetMemorysetByNameOrIdAnalysisByAnalysisJobIdParams(TypedDict):
     name_or_id: str
-    analysis_task_id: str
+    analysis_job_id: str
 
 
 class PostMemorysetByNameOrIdMemoryByMemoryIdCascadingEditsParams(TypedDict):
@@ -819,34 +901,50 @@ class DeleteFinetunedEmbeddingModelByNameOrIdParams(TypedDict):
     name_or_id: str
 
 
-class PostFinetunedEmbeddingModelByNameOrIdEvaluationParams(TypedDict):
+class PostGpuFinetunedEmbeddingModelByNameOrIdEmbeddingParams(TypedDict):
     name_or_id: str
 
 
-class GetFinetunedEmbeddingModelByNameOrIdEvaluationByTaskIdParams(TypedDict):
-    name_or_id: str
-    task_id: str
+class GetPretrainedEmbeddingModelByModelNameParams(TypedDict):
+    model_name: PretrainedEmbeddingModelName
 
 
-class GetFinetunedEmbeddingModelByNameOrIdEvaluationsParams(TypedDict):
-    name_or_id: str
+class PostGpuPretrainedEmbeddingModelByModelNameEmbeddingParams(TypedDict):
+    model_name: PretrainedEmbeddingModelName
 
 
-class GetPretrainedEmbeddingModelByModelNameParams(TypedDict):
-    model_name: PretrainedEmbeddingModelName
+class PostFinetunedEmbeddingModelByNameOrIdEvaluationParams(TypedDict):
+    name_or_id: str
 
 
 class PostPretrainedEmbeddingModelByModelNameEvaluationParams(TypedDict):
     model_name: PretrainedEmbeddingModelName
 
 
-class GetPretrainedEmbeddingModelByModelNameEvaluationByTaskIdParams(TypedDict):
+class GetFinetunedEmbeddingModelByNameOrIdEvaluationByJobIdParams(TypedDict):
+    name_or_id: str
+    job_id: str
+
+
+class GetPretrainedEmbeddingModelByModelNameEvaluationByJobIdParams(TypedDict):
     model_name: PretrainedEmbeddingModelName
-    task_id: str
+    job_id: str
+
+
+class GetFinetunedEmbeddingModelByNameOrIdEvaluationsParams(TypedDict):
+    name_or_id: str
+    datasource: NotRequired[str | None]
+    value_column: NotRequired[str | None]
+    label_column: NotRequired[str | None]
+    score_column: NotRequired[str | None]
 
 
 class GetPretrainedEmbeddingModelByModelNameEvaluationsParams(TypedDict):
     model_name: PretrainedEmbeddingModelName
+    datasource: NotRequired[str | None]
+    value_column: NotRequired[str | None]
+    label_column: NotRequired[str | None]
+    score_column: NotRequired[str | None]
 
 
 class PostDatasourceUploadRequest(TypedDict):
@@ -868,20 +966,19 @@ class DeleteDatasourceByNameOrIdParams(TypedDict):
     name_or_id: str
 
 
-class PostDatasourceByNameOrIdEmbeddingEvaluationParams(TypedDict):
+class PostDatasourceByNameOrIdRowsParams(TypedDict):
     name_or_id: str
 
 
-class GetDatasourceByNameOrIdEmbeddingEvaluationParams(TypedDict):
+class PostDatasourceByNameOrIdRowsCountParams(TypedDict):
     name_or_id: str
-    status: NotRequired[TaskStatus | None]
-    limit: NotRequired[int | None]
-    offset: NotRequired[int | None]
 
 
-class GetDatasourceByNameOrIdEmbeddingEvaluationByTaskIdParams(TypedDict):
+class GetDatasourceByNameOrIdEmbeddingModelEvaluationsParams(TypedDict):
     name_or_id: str
-    task_id: str
+    value_column: NotRequired[str | None]
+    label_column: NotRequired[str | None]
+    score_column: NotRequired[str | None]
 
 
 class GetDatasourceByNameOrIdDownloadParams(TypedDict):
@@ -907,36 +1004,42 @@ class DeleteClassificationModelByNameOrIdParams(TypedDict):
     name_or_id: str
 
 
-class PostClassificationModelByModelNameOrIdEvaluationParams(TypedDict):
-    model_name_or_id: str
+class PatchRegressionModelByNameOrIdParams(TypedDict):
+    name_or_id: str
 
 
-class GetClassificationModelByModelNameOrIdEvaluationParams(TypedDict):
-    model_name_or_id: str
+class GetRegressionModelByNameOrIdParams(TypedDict):
+    name_or_id: str
 
 
-class GetClassificationModelByModelNameOrIdEvaluationByTaskIdParams(TypedDict):
-    model_name_or_id: str
-    task_id: str
+class DeleteRegressionModelByNameOrIdParams(TypedDict):
+    name_or_id: str
 
 
-class DeleteClassificationModelByModelNameOrIdEvaluationByTaskIdParams(TypedDict):
-    model_name_or_id: str
-    task_id: str
+class PostGpuClassificationModelByNameOrIdPredictionParams(TypedDict):
+    name_or_id: str
 
 
-class PatchRegressionModelByNameOrIdParams(TypedDict):
+class PostClassificationModelByNameOrIdPredictionParams(TypedDict):
     name_or_id: str
 
 
-class GetRegressionModelByNameOrIdParams(TypedDict):
+class PostGpuRegressionModelByNameOrIdPredictionParams(TypedDict):
     name_or_id: str
 
 
-class DeleteRegressionModelByNameOrIdParams(TypedDict):
+class PostRegressionModelByNameOrIdPredictionParams(TypedDict):
     name_or_id: str
 
 
+class PostClassificationModelByModelNameOrIdEvaluationParams(TypedDict):
+    model_name_or_id: str
+
+
+class GetClassificationModelByModelNameOrIdEvaluationParams(TypedDict):
+    model_name_or_id: str
+
+
 class PostRegressionModelByModelNameOrIdEvaluationParams(TypedDict):
     model_name_or_id: str
 
@@ -945,26 +1048,36 @@ class GetRegressionModelByModelNameOrIdEvaluationParams(TypedDict):
     model_name_or_id: str
 
 
-class GetRegressionModelByModelNameOrIdEvaluationByTaskIdParams(TypedDict):
+class GetClassificationModelByModelNameOrIdEvaluationByJobIdParams(TypedDict):
     model_name_or_id: str
-    task_id: str
+    job_id: str
 
 
-class DeleteRegressionModelByModelNameOrIdEvaluationByTaskIdParams(TypedDict):
+class DeleteClassificationModelByModelNameOrIdEvaluationByJobIdParams(TypedDict):
     model_name_or_id: str
-    task_id: str
+    job_id: str
 
 
-class GetTaskByTaskIdParams(TypedDict):
-    task_id: str
+class GetRegressionModelByModelNameOrIdEvaluationByJobIdParams(TypedDict):
+    model_name_or_id: str
+    job_id: str
 
 
-class GetTaskByTaskIdStatusParams(TypedDict):
-    task_id: str
+class DeleteRegressionModelByModelNameOrIdEvaluationByJobIdParams(TypedDict):
+    model_name_or_id: str
+    job_id: str
+
+
+class GetJobByJobIdParams(TypedDict):
+    job_id: str
 
 
-class GetTaskParams(TypedDict):
-    status: NotRequired[TaskStatus | list[TaskStatus] | None]
+class GetJobByJobIdStatusParams(TypedDict):
+    job_id: str
+
+
+class GetJobParams(TypedDict):
+    status: NotRequired[JobStatus | list[JobStatus] | None]
     type: NotRequired[str | list[str] | None]
     limit: NotRequired[int | None]
     offset: NotRequired[int]
@@ -972,8 +1085,18 @@ class GetTaskParams(TypedDict):
     end_timestamp: NotRequired[str | None]
 
 
-class DeleteTaskByTaskIdAbortParams(TypedDict):
-    task_id: str
+class DeleteJobByJobIdAbortParams(TypedDict):
+    job_id: str
+
+
+class GetWorkerParams(TypedDict):
+    status: NotRequired[WorkerStatus | list[WorkerStatus] | None]
+    limit: NotRequired[int | None]
+    offset: NotRequired[int]
+
+
+class GetWorkerByWorkerIdParams(TypedDict):
+    worker_id: str
 
 
 class GetTelemetryPredictionByPredictionIdParams(TypedDict):
@@ -1019,43 +1142,8 @@ class DeleteTelemetryFeedbackCategoryByNameOrIdParams(TypedDict):
 PutTelemetryPredictionFeedbackRequest = list[PredictionFeedbackRequest]
 
 
-class GetAgentsBootstrapClassificationModelByTaskIdParams(TypedDict):
-    task_id: str
-
-
-class PostGpuMemorysetByNameOrIdLookupParams(TypedDict):
-    name_or_id: str
-
-
-class PatchGpuMemorysetByNameOrIdMemoryParams(TypedDict):
-    name_or_id: str
-
-
-class PostGpuMemorysetByNameOrIdMemoryParams(TypedDict):
-    name_or_id: str
-
-
-PostGpuMemorysetByNameOrIdMemoryRequest = list[LabeledMemoryInsert] | list[ScoredMemoryInsert]
-
-
-class PatchGpuMemorysetByNameOrIdMemoriesParams(TypedDict):
-    name_or_id: str
-
-
-class PostGpuClassificationModelByNameOrIdPredictionParams(TypedDict):
-    name_or_id: str
-
-
-class PostGpuRegressionModelByNameOrIdPredictionParams(TypedDict):
-    name_or_id: str
-
-
-class PostGpuFinetunedEmbeddingModelByNameOrIdEmbeddingParams(TypedDict):
-    name_or_id: str
-
-
-class PostGpuPretrainedEmbeddingModelByModelNameEmbeddingParams(TypedDict):
-    model_name: PretrainedEmbeddingModelName
+class GetAgentsBootstrapClassificationModelByJobIdParams(TypedDict):
+    job_id: str
 
 
 class FieldValidationError(TypedDict):
@@ -1127,6 +1215,7 @@ class ClassificationPredictionRequest(TypedDict):
     prompt: NotRequired[str | None]
     use_lookup_cache: NotRequired[bool]
     consistency_level: NotRequired[Literal["Bounded", "Session", "Strong", "Eventual"] | None]
+    ignore_unlabeled: NotRequired[bool]
 
 
 class CloneMemorysetRequest(TypedDict):
@@ -1143,6 +1232,7 @@ class ColumnInfo(TypedDict):
     name: str
     type: ColumnType
     enum_options: NotRequired[list[str] | None]
+    string_values: NotRequired[list[str] | None]
     int_values: NotRequired[list[int] | None]
     contains_nones: NotRequired[bool]
 
@@ -1154,6 +1244,7 @@ class ConceptMetrics(TypedDict):
     primary_label: int | None
     memory_count: int
     subconcepts: list[SubConceptMetrics]
+    label_percentages: list[LabelPercentage]
 
 
 class CreateClassificationModelRequest(TypedDict):
@@ -1187,6 +1278,9 @@ class CreateMemorysetRequest(TypedDict):
     index_params: NotRequired[dict[str, int | float | str]]
     prompt: NotRequired[str]
     hidden: NotRequired[bool]
+    batch_size: NotRequired[int]
+    subsample: NotRequired[int | float | None]
+    memory_type: NotRequired[MemoryType]
 
 
 class CreateRegressionModelRequest(TypedDict):
@@ -1198,25 +1292,6 @@ class CreateRegressionModelRequest(TypedDict):
     head_type: NotRequired[RARHeadType]
 
 
-class DatasourceEmbeddingEvaluationsRequest(TypedDict):
-    value_column: str
-    label_column: str
-    source_id_column: str | None
-    neighbor_count: NotRequired[int]
-    label_names: NotRequired[list[str] | None]
-    embedding_models: NotRequired[list[PretrainedEmbeddingModelName | str] | None]
-
-
-class DatasourceEmbeddingEvaluationsTaskPayload(TypedDict):
-    value_column: str
-    label_column: str
-    source_id_column: str | None
-    neighbor_count: NotRequired[int]
-    label_names: NotRequired[list[str] | None]
-    embedding_models: NotRequired[list[PretrainedEmbeddingModelName | str] | None]
-    datasource_id: str
-
-
 class DatasourceMetadata(TypedDict):
     id: str
     org_id: str
@@ -1230,56 +1305,52 @@ class DatasourceMetadata(TypedDict):
 
 
 class EmbeddingEvaluationResponse(TypedDict):
-    task_id: str
+    job_id: str
     org_id: str
-    status: TaskStatus
+    finetuned_embedding_model_id: str | None
+    pretrained_embedding_model_name: PretrainedEmbeddingModelName | None
+    datasource_id: str
+    subsample: int | float | None
+    datasource_value_column: str
+    datasource_label_column: NotRequired[str | None]
+    datasource_score_column: NotRequired[str | None]
+    neighbor_count: int
+    weigh_memories: bool
+    status: JobStatus
     result: ClassificationMetrics | RegressionMetrics | None
     created_at: str
     updated_at: str
-
-
-class EmbeddingEvaluationResponseUnionClassificationMetricsRegressionMetrics(TypedDict):
     task_id: str
-    org_id: str
-    status: TaskStatus
-    result: ClassificationMetrics | RegressionMetrics | None
-    created_at: str
-    updated_at: str
-
-
-class EmbeddingModelResult(TypedDict):
-    embedding_model_name: str
-    embedding_model_path: str
-    analysis_result: MemorysetLabelMetrics
-    memoryset_name: NotRequired[str | None]
-    is_finetuned: NotRequired[bool]
 
 
 class EvaluationResponse(TypedDict):
-    task_id: str
+    job_id: str
     org_id: str
-    status: TaskStatus
+    status: JobStatus
     result: ClassificationMetrics | RegressionMetrics | None
     created_at: str
     updated_at: str
+    task_id: str
 
 
 class EvaluationResponseClassificationMetrics(TypedDict):
-    task_id: str
+    job_id: str
     org_id: str
-    status: TaskStatus
+    status: JobStatus
     result: ClassificationMetrics | None
     created_at: str
     updated_at: str
+    task_id: str
 
 
 class EvaluationResponseRegressionMetrics(TypedDict):
-    task_id: str
+    job_id: str
     org_id: str
-    status: TaskStatus
+    status: JobStatus
     result: RegressionMetrics | None
     created_at: str
     updated_at: str
+    task_id: str
 
 
 class FinetuneEmbeddingModelRequest(TypedDict):
@@ -1288,7 +1359,8 @@ class FinetuneEmbeddingModelRequest(TypedDict):
     train_memoryset_name_or_id: NotRequired[str | None]
     train_datasource_name_or_id: NotRequired[str | None]
     eval_datasource_name_or_id: NotRequired[str | None]
-    label_column: NotRequired[str]
+    label_column: NotRequired[str | None]
+    score_column: NotRequired[str | None]
     value_column: NotRequired[str]
     training_method: NotRequired[EmbeddingFinetuningMethod]
     training_args: NotRequired[dict[str, str | int | float | bool]]
@@ -1305,8 +1377,9 @@ class FinetunedEmbeddingModelMetadata(TypedDict):
     created_at: str
     updated_at: str
     base_model: PretrainedEmbeddingModelName
+    finetuning_job_id: str
+    finetuning_status: JobStatus
     finetuning_task_id: str
-    finetuning_status: TaskStatus
 
 
 class HTTPValidationError(TypedDict):
@@ -1318,10 +1391,28 @@ class InvalidInputErrorResponse(TypedDict):
     validation_issues: list[FieldValidationError]
 
 
+class Job(TypedDict):
+    status: JobStatus
+    steps_total: int | None
+    steps_completed: int | None
+    exception: str | None
+    updated_at: str
+    created_at: str
+    id: str
+    org_id: str
+    worker_id: str | None
+    type: str
+    payload: BaseModel
+    result: BaseModel | None
+    depends_on: NotRequired[list[str]]
+    lease_token: str | None
+
+
 class LabelPredictionMemoryLookup(TypedDict):
     value: str | bytes
     embedding: list[float]
     source_id: str | None
+    partition_id: str | None
     metadata: dict[str, str | int | float | bool | None]
     memory_id: str
     memory_version: int
@@ -1363,6 +1454,7 @@ class LabeledMemory(TypedDict):
     value: str | bytes
     embedding: list[float]
     source_id: str | None
+    partition_id: str | None
     metadata: dict[str, str | int | float | bool | None]
     memory_id: str
     memory_version: int
@@ -1378,6 +1470,7 @@ class LabeledMemoryLookup(TypedDict):
     value: str | bytes
     embedding: list[float]
     source_id: str | None
+    partition_id: str | None
     metadata: dict[str, str | int | float | bool | None]
     memory_id: str
     memory_version: int
@@ -1395,6 +1488,7 @@ class LabeledMemoryUpdate(TypedDict):
     value: NotRequired[str | bytes]
     metadata: NotRequired[dict[str, str | int | float | bool | None] | None]
     source_id: NotRequired[str | None]
+    partition_id: NotRequired[str | None]
     metrics: NotRequired[MemoryMetrics | None]
     label: NotRequired[int | None]
 
@@ -1403,6 +1497,7 @@ class LabeledMemoryWithFeedbackMetrics(TypedDict):
     value: str | bytes
     embedding: list[float]
     source_id: str | None
+    partition_id: str | None
     metadata: dict[str, str | int | float | bool | None]
     memory_id: str
     memory_version: int
@@ -1422,14 +1517,15 @@ class ListPredictionsRequest(TypedDict):
     prediction_ids: NotRequired[list[str] | None]
     start_timestamp: NotRequired[str | None]
     end_timestamp: NotRequired[str | None]
-    limit: NotRequired[int | None]
+    memory_id: NotRequired[str | None]
+    limit: NotRequired[int]
     offset: NotRequired[int | None]
     sort: NotRequired[PredictionSort]
     expected_label_match: NotRequired[bool | None]
 
 
 class MemorysetAnalysisConfigs(TypedDict):
-    neighbor: NotRequired[MemorysetNeighborAnalysisConfig | None]
+    distribution: NotRequired[MemorysetDistributionAnalysisConfig | None]
     label: NotRequired[MemorysetLabelAnalysisConfig | None]
     duplicate: NotRequired[MemorysetDuplicateAnalysisConfig | None]
     projection: NotRequired[MemorysetProjectionAnalysisConfig | None]
@@ -1452,7 +1548,7 @@ class MemorysetConceptMetrics(TypedDict):
 
 
 class MemorysetMetrics(TypedDict):
-    neighbor: NotRequired[MemorysetNeighborMetrics | None]
+    distribution: NotRequired[MemorysetDistributionMetrics | None]
     label: NotRequired[MemorysetLabelMetrics | None]
     duplicate: NotRequired[MemorysetDuplicateMetrics | None]
     projection: NotRequired[MemorysetProjectionMetrics | None]
@@ -1461,6 +1557,13 @@ class MemorysetMetrics(TypedDict):
     concepts: NotRequired[MemorysetConceptMetrics | None]
 
 
+class PaginatedJob(TypedDict):
+    items: list[Job]
+    total: int
+    offset: int
+    limit: int
+
+
 class PaginatedUnionLabeledMemoryWithFeedbackMetricsScoredMemoryWithFeedbackMetrics(TypedDict):
     items: list[LabeledMemoryWithFeedbackMetrics | ScoredMemoryWithFeedbackMetrics]
     total: int
@@ -1478,21 +1581,6 @@ class PretrainedEmbeddingModelMetadata(TypedDict):
     num_params: int
 
 
-class Task(TypedDict):
-    status: TaskStatus
-    steps_total: int | None
-    steps_completed: int | None
-    exception: str | None
-    updated_at: str
-    created_at: str
-    id: str
-    org_id: str
-    type: str
-    payload: BaseModel
-    result: BaseModel | None
-    depends_on: NotRequired[list[str]]
-
-
 class TelemetryMemoriesRequest(TypedDict):
     memoryset_id: str
     offset: NotRequired[int]
@@ -1501,6 +1589,16 @@ class TelemetryMemoriesRequest(TypedDict):
     sort: NotRequired[list[TelemetrySortOptions] | None]
 
 
+class WorkerInfo(TypedDict):
+    id: str
+    status: WorkerStatus
+    started_at: str
+    updated_at: str
+    version: str | None
+    git_sha: str
+    config: dict[str, str | float | int | bool | dict[str, str] | None]
+
+
 PatchGpuMemorysetByNameOrIdMemoryRequest = LabeledMemoryUpdate | ScoredMemoryUpdate
 
 
@@ -1513,15 +1611,11 @@ class CascadingEditSuggestion(TypedDict):
     lookup_score: float
 
 
-class EmbeddingEvaluationResult(TypedDict):
-    evaluation_results: list[EmbeddingModelResult]
-
-
 class MemorysetAnalysisResponse(TypedDict):
-    task_id: str
+    job_id: str
     org_id: str
     memoryset_id: str
-    status: TaskStatus
+    status: JobStatus
     lookup_count: int
     batch_size: int
     clear_metrics: bool
@@ -1529,6 +1623,7 @@ class MemorysetAnalysisResponse(TypedDict):
     results: MemorysetMetrics | None
     created_at: str
     updated_at: str
+    task_id: str
 
 
 class MemorysetMetadata(TypedDict):
@@ -1544,8 +1639,8 @@ class MemorysetMetadata(TypedDict):
     created_at: str
     updated_at: str
     memories_updated_at: str
-    insertion_task_id: str
-    insertion_status: TaskStatus
+    insertion_job_id: str
+    insertion_status: JobStatus
     metrics: MemorysetMetrics
     memory_type: MemoryType
     label_names: list[str] | None
@@ -1555,10 +1650,11 @@ class MemorysetMetadata(TypedDict):
     document_prompt_override: str | None
     query_prompt_override: str | None
     hidden: bool
+    insertion_task_id: str
 
 
-class PaginatedTask(TypedDict):
-    items: list[Task]
+class PaginatedWorkerInfo(TypedDict):
+    items: list[WorkerInfo]
     total: int
     offset: int
     limit: int
@@ -1572,22 +1668,12 @@ class BootstrapClassificationModelMeta(TypedDict):
 
 
 class BootstrapClassificationModelResponse(TypedDict):
-    task_id: str
+    job_id: str
     org_id: str
-    status: TaskStatus
+    status: JobStatus
     result: BootstrapClassificationModelMeta | None
     input: BootstrapClassificationModelRequest | None
-
-
-class DatasourceEmbeddingEvaluationsResponse(TypedDict):
-    task_id: str
-    org_id: str
-    datasource_id: str
-    status: TaskStatus
-    result: EmbeddingEvaluationResult | None
-    payload: DatasourceEmbeddingEvaluationsTaskPayload
-    created_at: str
-    updated_at: str
+    task_id: str
 
 
 class OrcaClient(Client):
@@ -1604,11 +1690,11 @@ class OrcaClient(Client):
         return path_params, query_params
 
     @overload
-    def DELETE(
+    def GET(
         self,
-        path: Literal["/cleanup"],
+        path: Literal["/test_error/{status_code}"],
         *,
-        params: None = None,
+        params: GetTestErrorByStatusCodeParams,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -1616,16 +1702,15 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> CleanupResponse:
-        """Cleanup orphaned milvus collections and blobs"""
+    ) -> Any:
         pass
 
     @overload
-    def DELETE(
+    def GET(
         self,
-        path: Literal["/auth/api_key/{name_or_id}"],
+        path: Literal["/check/alive"],
         *,
-        params: DeleteAuthApiKeyByNameOrIdParams,
+        params: None = None,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -1633,14 +1718,13 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> None:
-        """Delete an API key by name or ID."""
+    ) -> AliveResponse:
         pass
 
     @overload
-    def DELETE(
+    def GET(
         self,
-        path: Literal["/auth/org"],
+        path: Literal["/check/ready"],
         *,
         params: None = None,
         parse_as: Literal["json"] = "json",
@@ -1650,16 +1734,15 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> None:
-        """Deletes the org and all associated resources"""
+    ) -> ReadyResponse:
         pass
 
     @overload
-    def DELETE(
+    def GET(
         self,
-        path: Literal["/memoryset/{name_or_id}"],
+        path: Literal["/gpu/check/healthy"],
         *,
-        params: DeleteMemorysetByNameOrIdParams,
+        params: None = None,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -1667,15 +1750,15 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> None:
+    ) -> HealthyResponse:
         pass
 
     @overload
-    def DELETE(
+    def GET(
         self,
-        path: Literal["/memoryset/{name_or_id}/memory/{memory_id}"],
+        path: Literal["/check/healthy"],
         *,
-        params: DeleteMemorysetByNameOrIdMemoryByMemoryIdParams,
+        params: None = None,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -1683,15 +1766,15 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> None:
+    ) -> HealthyResponse:
         pass
 
     @overload
-    def DELETE(
+    def GET(
         self,
-        path: Literal["/finetuned_embedding_model/{name_or_id}"],
+        path: Literal["/gpu/config"],
         *,
-        params: DeleteFinetunedEmbeddingModelByNameOrIdParams,
+        params: None = None,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -1699,16 +1782,15 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> None:
-        """Delete a finetuned embedding model by name or ID."""
+    ) -> dict[str, str | float | int | bool | None]:
         pass
 
     @overload
-    def DELETE(
+    def GET(
         self,
-        path: Literal["/datasource/{name_or_id}"],
+        path: Literal["/config"],
         *,
-        params: DeleteDatasourceByNameOrIdParams,
+        params: None = None,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -1716,16 +1798,15 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> None:
-        """Delete a datasource by name or ID."""
+    ) -> dict[str, str | float | int | bool | None]:
         pass
 
     @overload
-    def DELETE(
+    def GET(
         self,
-        path: Literal["/classification_model/{name_or_id}"],
+        path: Literal["/auth/root"],
         *,
-        params: DeleteClassificationModelByNameOrIdParams,
+        params: None = None,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -1733,15 +1814,16 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> None:
+    ) -> bool:
+        """Return true only when called with a valid root API key; otherwise 401 Unauthenticated."""
         pass
 
     @overload
-    def DELETE(
+    def GET(
         self,
-        path: Literal["/classification_model/{model_name_or_id}/evaluation/{task_id}"],
+        path: Literal["/auth/api_key"],
         *,
-        params: DeleteClassificationModelByModelNameOrIdEvaluationByTaskIdParams,
+        params: None = None,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -1749,15 +1831,16 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> None:
+    ) -> list[ApiKeyMetadata]:
+        """List all API keys for the organization."""
         pass
 
     @overload
-    def DELETE(
+    def GET(
         self,
-        path: Literal["/regression_model/{name_or_id}"],
+        path: Literal["/auth"],
         *,
-        params: DeleteRegressionModelByNameOrIdParams,
+        params: None = None,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -1765,15 +1848,16 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> None:
+    ) -> bool:
+        """Returns true if the api key header is valid for the org (will be false for admin api key)"""
         pass
 
     @overload
-    def DELETE(
+    def GET(
         self,
-        path: Literal["/regression_model/{model_name_or_id}/evaluation/{task_id}"],
+        path: Literal["/auth/org/plan"],
         *,
-        params: DeleteRegressionModelByModelNameOrIdEvaluationByTaskIdParams,
+        params: None = None,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -1781,15 +1865,16 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> None:
+    ) -> OrgPlan:
+        """Get the organization plan."""
         pass
 
     @overload
-    def DELETE(
+    def GET(
         self,
-        path: Literal["/task/{task_id}/abort"],
+        path: Literal["/memoryset"],
         *,
-        params: DeleteTaskByTaskIdAbortParams,
+        params: GetMemorysetParams,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -1797,15 +1882,15 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> None:
+    ) -> list[MemorysetMetadata]:
         pass
 
     @overload
-    def DELETE(
+    def GET(
         self,
-        path: Literal["/telemetry/feedback_category/{name_or_id}"],
+        path: Literal["/memoryset/{name_or_id}"],
         *,
-        params: DeleteTelemetryFeedbackCategoryByNameOrIdParams,
+        params: GetMemorysetByNameOrIdParams,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -1813,46 +1898,31 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> None:
-        """Delete a feedback category and all associated feedback records."""
+    ) -> MemorysetMetadata:
         pass
 
-    def DELETE(
+    @overload
+    def GET(
         self,
-        path: str,
+        path: Literal["/memoryset/{name_or_id}/memory/{memory_id}"],
         *,
-        params: Mapping[str, Any] | None = None,
-        parse_as: Literal["json", "text"] | None = "json",
+        params: GetMemorysetByNameOrIdMemoryByMemoryIdParams,
+        parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
         auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> Any:
-        path_params, query_params = self._parse_params(params or {}, path)
-        res = self.delete(
-            path.format(**path_params),
-            params=query_params,
-            headers=headers,
-            cookies=cookies,
-            auth=auth,
-            follow_redirects=follow_redirects,
-            timeout=timeout,
-            extensions=extensions,
-        ).raise_for_status()
-        return (
-            None
-            if res.status_code == 204
-            else res.json() if parse_as == "json" else res.text if parse_as == "text" else res.content
-        )
+    ) -> LabeledMemory | ScoredMemory:
+        pass
 
     @overload
     def GET(
         self,
-        path: Literal["/check/alive"],
+        path: Literal["/memoryset/{name_or_id}/potential_duplicate_groups"],
         *,
-        params: None = None,
+        params: GetMemorysetByNameOrIdPotentialDuplicateGroupsParams,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -1860,15 +1930,15 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> AliveResponse:
+    ) -> list[list[LabeledMemory]] | list[list[ScoredMemory]]:
         pass
 
     @overload
     def GET(
         self,
-        path: Literal["/check/ready"],
+        path: Literal["/memoryset/{name_or_id}/analysis"],
         *,
-        params: None = None,
+        params: GetMemorysetByNameOrIdAnalysisParams,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -1876,15 +1946,15 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> ReadyResponse:
+    ) -> list[MemorysetAnalysisResponse]:
         pass
 
     @overload
     def GET(
         self,
-        path: Literal["/gpu/check/healthy"],
+        path: Literal["/memoryset/{name_or_id}/analysis/{analysis_job_id}"],
         *,
-        params: None = None,
+        params: GetMemorysetByNameOrIdAnalysisByAnalysisJobIdParams,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -1892,13 +1962,13 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> HealthyResponse:
+    ) -> MemorysetAnalysisResponse:
         pass
 
     @overload
     def GET(
         self,
-        path: Literal["/check/healthy"],
+        path: Literal["/finetuned_embedding_model"],
         *,
         params: None = None,
         parse_as: Literal["json"] = "json",
@@ -1908,15 +1978,16 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> HealthyResponse:
+    ) -> list[FinetunedEmbeddingModelMetadata]:
+        """List all finetuned embedding models for the organization."""
         pass
 
     @overload
     def GET(
         self,
-        path: Literal["/gpu/config"],
+        path: Literal["/finetuned_embedding_model/{name_or_id}"],
         *,
-        params: None = None,
+        params: GetFinetunedEmbeddingModelByNameOrIdParams,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -1924,13 +1995,14 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> dict[str, str | float | int | bool | None]:
+    ) -> FinetunedEmbeddingModelMetadata:
+        """Get a finetuned embedding model by name or ID."""
         pass
 
     @overload
     def GET(
         self,
-        path: Literal["/config"],
+        path: Literal["/pretrained_embedding_model"],
         *,
         params: None = None,
         parse_as: Literal["json"] = "json",
@@ -1940,47 +2012,50 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> dict[str, str | float | int | bool | None]:
+    ) -> list[PretrainedEmbeddingModelMetadata]:
+        """List all available pretrained embedding models."""
         pass
 
     @overload
     def GET(
         self,
-        path: Literal["/gpu/"],
+        path: Literal["/pretrained_embedding_model/{model_name}"],
         *,
-        params: None = None,
-        parse_as: Literal["text"],
+        params: GetPretrainedEmbeddingModelByModelNameParams,
+        parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
         auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> str:
+    ) -> PretrainedEmbeddingModelMetadata:
+        """Get metadata for a specific pretrained embedding model."""
         pass
 
     @overload
     def GET(
         self,
-        path: Literal["/"],
+        path: Literal["/finetuned_embedding_model/{name_or_id}/evaluation/{job_id}"],
         *,
-        params: None = None,
-        parse_as: Literal["text"],
+        params: GetFinetunedEmbeddingModelByNameOrIdEvaluationByJobIdParams,
+        parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
         auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> str:
+    ) -> EmbeddingEvaluationResponse:
+        """Get evaluation results for a finetuned embedding model by job ID."""
         pass
 
     @overload
     def GET(
         self,
-        path: Literal["/auth/root"],
+        path: Literal["/pretrained_embedding_model/{model_name}/evaluation/{job_id}"],
         *,
-        params: None = None,
+        params: GetPretrainedEmbeddingModelByModelNameEvaluationByJobIdParams,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -1988,16 +2063,16 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> bool:
-        """Return true only when called with a valid root API key; otherwise 401 Unauthenticated."""
+    ) -> EmbeddingEvaluationResponse:
+        """Get evaluation results for a pretrained embedding model by job ID."""
         pass
 
     @overload
     def GET(
         self,
-        path: Literal["/auth/api_key"],
+        path: Literal["/finetuned_embedding_model/{name_or_id}/evaluations"],
         *,
-        params: None = None,
+        params: GetFinetunedEmbeddingModelByNameOrIdEvaluationsParams,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -2005,16 +2080,16 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> list[ApiKeyMetadata]:
-        """List all API keys for the organization."""
+    ) -> list[EmbeddingEvaluationResponse]:
+        """List all evaluation results for a finetuned embedding model."""
         pass
 
     @overload
     def GET(
         self,
-        path: Literal["/auth"],
+        path: Literal["/pretrained_embedding_model/{model_name}/evaluations"],
         *,
-        params: None = None,
+        params: GetPretrainedEmbeddingModelByModelNameEvaluationsParams,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -2022,14 +2097,14 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> bool:
-        """Returns true if the api key header is valid for the org (will be false for admin api key)"""
+    ) -> list[EmbeddingEvaluationResponse]:
+        """List all evaluation results for a pretrained embedding model."""
         pass
 
     @overload
     def GET(
         self,
-        path: Literal["/auth/org/plan"],
+        path: Literal["/datasource"],
         *,
         params: None = None,
         parse_as: Literal["json"] = "json",
@@ -2039,16 +2114,16 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> OrgPlan:
-        """Get the organization plan."""
+    ) -> list[DatasourceMetadata]:
+        """List all datasources for the organization."""
         pass
 
     @overload
     def GET(
         self,
-        path: Literal["/memoryset"],
+        path: Literal["/datasource/{name_or_id}"],
         *,
-        params: GetMemorysetParams,
+        params: GetDatasourceByNameOrIdParams,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -2056,15 +2131,16 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> list[MemorysetMetadata]:
+    ) -> DatasourceMetadata:
+        """Get a datasource by name or ID."""
         pass
 
     @overload
     def GET(
         self,
-        path: Literal["/memoryset/{name_or_id}"],
+        path: Literal["/datasource/{name_or_id}/embedding_model_evaluations"],
         *,
-        params: GetMemorysetByNameOrIdParams,
+        params: GetDatasourceByNameOrIdEmbeddingModelEvaluationsParams,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -2072,15 +2148,16 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> MemorysetMetadata:
+    ) -> list[EmbeddingEvaluationResponse]:
+        """List all evaluation results for a datasource."""
         pass
 
     @overload
     def GET(
         self,
-        path: Literal["/memoryset/{name_or_id}/memory/{memory_id}"],
+        path: Literal["/datasource/{name_or_id}/download"],
         *,
-        params: GetMemorysetByNameOrIdMemoryByMemoryIdParams,
+        params: GetDatasourceByNameOrIdDownloadParams,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -2088,47 +2165,50 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> LabeledMemory | ScoredMemory:
+    ) -> list[dict[str, Any]]:
+        """Download datasource in the specified format."""
         pass
 
     @overload
     def GET(
         self,
-        path: Literal["/memoryset/{name_or_id}/potential_duplicate_groups"],
+        path: Literal["/datasource/{name_or_id}/download"],
         *,
-        params: GetMemorysetByNameOrIdPotentialDuplicateGroupsParams,
-        parse_as: Literal["json"] = "json",
+        params: GetDatasourceByNameOrIdDownloadParams,
+        parse_as: Literal["text"],
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
         auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> list[list[LabeledMemory]] | list[list[ScoredMemory]]:
+    ) -> str:
+        """Download datasource in the specified format."""
         pass
 
     @overload
     def GET(
         self,
-        path: Literal["/memoryset/{name_or_id}/analysis"],
+        path: Literal["/datasource/{name_or_id}/download"],
         *,
-        params: GetMemorysetByNameOrIdAnalysisParams,
-        parse_as: Literal["json"] = "json",
+        params: GetDatasourceByNameOrIdDownloadParams,
+        parse_as: None,
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
         auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> list[MemorysetAnalysisResponse]:
+    ) -> bytes:
+        """Download datasource in the specified format."""
         pass
 
     @overload
     def GET(
         self,
-        path: Literal["/memoryset/{name_or_id}/analysis/{analysis_task_id}"],
+        path: Literal["/classification_model"],
         *,
-        params: GetMemorysetByNameOrIdAnalysisByAnalysisTaskIdParams,
+        params: None = None,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -2136,13 +2216,13 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> MemorysetAnalysisResponse:
+    ) -> list[ClassificationModelMetadata]:
         pass
 
     @overload
     def GET(
         self,
-        path: Literal["/finetuned_embedding_model"],
+        path: Literal["/regression_model"],
         *,
         params: None = None,
         parse_as: Literal["json"] = "json",
@@ -2152,16 +2232,15 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> list[FinetunedEmbeddingModelMetadata]:
-        """List all finetuned embedding models for the organization."""
+    ) -> list[RegressionModelMetadata]:
         pass
 
     @overload
     def GET(
         self,
-        path: Literal["/finetuned_embedding_model/{name_or_id}"],
+        path: Literal["/classification_model/{name_or_id}"],
         *,
-        params: GetFinetunedEmbeddingModelByNameOrIdParams,
+        params: GetClassificationModelByNameOrIdParams,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -2169,16 +2248,15 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> FinetunedEmbeddingModelMetadata:
-        """Get a finetuned embedding model by name or ID."""
+    ) -> ClassificationModelMetadata:
         pass
 
     @overload
     def GET(
         self,
-        path: Literal["/finetuned_embedding_model/{name_or_id}/evaluation/{task_id}"],
+        path: Literal["/regression_model/{name_or_id}"],
         *,
-        params: GetFinetunedEmbeddingModelByNameOrIdEvaluationByTaskIdParams,
+        params: GetRegressionModelByNameOrIdParams,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -2186,16 +2264,15 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> EmbeddingEvaluationResponseUnionClassificationMetricsRegressionMetrics:
-        """Get evaluation results for a finetuned embedding model by task ID."""
+    ) -> RegressionModelMetadata:
         pass
 
     @overload
     def GET(
         self,
-        path: Literal["/finetuned_embedding_model/{name_or_id}/evaluations"],
+        path: Literal["/predictive_model"],
         *,
-        params: GetFinetunedEmbeddingModelByNameOrIdEvaluationsParams,
+        params: None = None,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -2203,16 +2280,15 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> list[EmbeddingEvaluationResponseUnionClassificationMetricsRegressionMetrics]:
-        """List all evaluation results for a finetuned embedding model."""
+    ) -> list[ClassificationModelMetadata | RegressionModelMetadata]:
         pass
 
     @overload
     def GET(
         self,
-        path: Literal["/pretrained_embedding_model"],
+        path: Literal["/classification_model/{model_name_or_id}/evaluation"],
         *,
-        params: None = None,
+        params: GetClassificationModelByModelNameOrIdEvaluationParams,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -2220,16 +2296,15 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> list[PretrainedEmbeddingModelMetadata]:
-        """List all available pretrained embedding models."""
+    ) -> list[EvaluationResponseClassificationMetrics]:
         pass
 
     @overload
     def GET(
         self,
-        path: Literal["/pretrained_embedding_model/{model_name}"],
+        path: Literal["/regression_model/{model_name_or_id}/evaluation"],
         *,
-        params: GetPretrainedEmbeddingModelByModelNameParams,
+        params: GetRegressionModelByModelNameOrIdEvaluationParams,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -2237,16 +2312,15 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> PretrainedEmbeddingModelMetadata:
-        """Get metadata for a specific pretrained embedding model."""
+    ) -> list[EvaluationResponseRegressionMetrics]:
         pass
 
     @overload
     def GET(
         self,
-        path: Literal["/pretrained_embedding_model/{model_name}/evaluation/{task_id}"],
+        path: Literal["/classification_model/{model_name_or_id}/evaluation/{job_id}"],
         *,
-        params: GetPretrainedEmbeddingModelByModelNameEvaluationByTaskIdParams,
+        params: GetClassificationModelByModelNameOrIdEvaluationByJobIdParams,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -2254,16 +2328,15 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> EmbeddingEvaluationResponseUnionClassificationMetricsRegressionMetrics:
-        """Get evaluation results for a pretrained embedding model by task ID."""
+    ) -> EvaluationResponseClassificationMetrics:
         pass
 
     @overload
     def GET(
         self,
-        path: Literal["/pretrained_embedding_model/{model_name}/evaluations"],
+        path: Literal["/regression_model/{model_name_or_id}/evaluation/{job_id}"],
         *,
-        params: GetPretrainedEmbeddingModelByModelNameEvaluationsParams,
+        params: GetRegressionModelByModelNameOrIdEvaluationByJobIdParams,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -2271,16 +2344,15 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> list[EmbeddingEvaluationResponseUnionClassificationMetricsRegressionMetrics]:
-        """List all evaluation results for a pretrained embedding model."""
+    ) -> EvaluationResponseRegressionMetrics:
         pass
 
     @overload
     def GET(
         self,
-        path: Literal["/datasource"],
+        path: Literal["/job/{job_id}"],
         *,
-        params: None = None,
+        params: GetJobByJobIdParams,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -2288,16 +2360,15 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> list[DatasourceMetadata]:
-        """List all datasources for the organization."""
+    ) -> Job:
         pass
 
     @overload
     def GET(
         self,
-        path: Literal["/datasource/{name_or_id}"],
+        path: Literal["/job/{job_id}/status"],
         *,
-        params: GetDatasourceByNameOrIdParams,
+        params: GetJobByJobIdStatusParams,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -2305,16 +2376,15 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> DatasourceMetadata:
-        """Get a datasource by name or ID."""
+    ) -> JobStatusInfo:
         pass
 
     @overload
     def GET(
         self,
-        path: Literal["/datasource/{name_or_id}/embedding_evaluation"],
+        path: Literal["/job"],
         *,
-        params: GetDatasourceByNameOrIdEmbeddingEvaluationParams,
+        params: GetJobParams,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -2322,16 +2392,15 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> list[DatasourceEmbeddingEvaluationsResponse]:
-        """List embedding evaluation tasks for a datasource."""
+    ) -> PaginatedJob:
         pass
 
     @overload
     def GET(
         self,
-        path: Literal["/datasource/{name_or_id}/embedding_evaluation/{task_id}"],
+        path: Literal["/worker"],
         *,
-        params: GetDatasourceByNameOrIdEmbeddingEvaluationByTaskIdParams,
+        params: GetWorkerParams,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -2339,16 +2408,20 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> DatasourceEmbeddingEvaluationsResponse:
-        """Get an embedding evaluation task by ID."""
+    ) -> PaginatedWorkerInfo:
+        """
+        List all workers in the system. Requires root access.
+
+        This endpoint automatically cleans up orphaned workers before returning results.
+        """
         pass
 
     @overload
     def GET(
         self,
-        path: Literal["/datasource/{name_or_id}/download"],
+        path: Literal["/worker/{worker_id}"],
         *,
-        params: GetDatasourceByNameOrIdDownloadParams,
+        params: GetWorkerByWorkerIdParams,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -2356,50 +2429,50 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> list[dict[str, Any]]:
-        """Download datasource in the specified format."""
+    ) -> WorkerInfo:
+        """Get information about a specific worker. Requires root access."""
         pass
 
     @overload
     def GET(
         self,
-        path: Literal["/datasource/{name_or_id}/download"],
+        path: Literal["/telemetry/prediction/{prediction_id}"],
         *,
-        params: GetDatasourceByNameOrIdDownloadParams,
-        parse_as: Literal["text"],
+        params: GetTelemetryPredictionByPredictionIdParams,
+        parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
         auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> str:
-        """Download datasource in the specified format."""
+    ) -> LabelPredictionWithMemoriesAndFeedback | ScorePredictionWithMemoriesAndFeedback:
+        """Get a specific prediction by ID."""
         pass
 
     @overload
     def GET(
         self,
-        path: Literal["/datasource/{name_or_id}/download"],
+        path: Literal["/telemetry/prediction/{prediction_id}/explanation"],
         *,
-        params: GetDatasourceByNameOrIdDownloadParams,
-        parse_as: None,
+        params: GetTelemetryPredictionByPredictionIdExplanationParams,
+        parse_as: Literal["text"],
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
         auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> bytes:
-        """Download datasource in the specified format."""
+    ) -> str:
+        """Get explanation for a prediction, optionally streaming the response."""
         pass
 
     @overload
     def GET(
         self,
-        path: Literal["/predictive_model"],
+        path: Literal["/telemetry/prediction/{prediction_id}/action"],
         *,
-        params: None = None,
+        params: GetTelemetryPredictionByPredictionIdActionParams,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -2407,15 +2480,16 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> list[ClassificationModelMetadata | RegressionModelMetadata]:
+    ) -> ActionRecommendation:
+        """Get action recommendation for improving a specific prediction."""
         pass
 
     @overload
     def GET(
         self,
-        path: Literal["/classification_model"],
+        path: Literal["/telemetry/prediction/{prediction_id}/memory_suggestions"],
         *,
-        params: None = None,
+        params: GetTelemetryPredictionByPredictionIdMemorySuggestionsParams,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -2423,15 +2497,21 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> list[ClassificationModelMetadata]:
+    ) -> AddMemoryRecommendations:
+        """
+        Generate synthetic memory suggestions to improve a specific prediction.
+
+        The returned suggestions have labels as string representations of integer indices
+        corresponding to the memoryset's label_names.
+        """
         pass
 
     @overload
     def GET(
         self,
-        path: Literal["/classification_model/{name_or_id}"],
+        path: Literal["/telemetry/feedback_category"],
         *,
-        params: GetClassificationModelByNameOrIdParams,
+        params: None = None,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -2439,15 +2519,16 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> ClassificationModelMetadata:
+    ) -> list[PredictionFeedbackCategory]:
+        """List all feedback categories for the organization."""
         pass
 
     @overload
     def GET(
         self,
-        path: Literal["/classification_model/{model_name_or_id}/evaluation"],
+        path: Literal["/telemetry/feedback_category/{name_or_id}"],
         *,
-        params: GetClassificationModelByModelNameOrIdEvaluationParams,
+        params: GetTelemetryFeedbackCategoryByNameOrIdParams,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -2455,15 +2536,16 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> list[EvaluationResponseClassificationMetrics]:
+    ) -> PredictionFeedbackCategory:
+        """Get a feedback category by name or ID."""
         pass
 
     @overload
     def GET(
         self,
-        path: Literal["/classification_model/{model_name_or_id}/evaluation/{task_id}"],
+        path: Literal["/agents/bootstrap_classification_model/{job_id}"],
         *,
-        params: GetClassificationModelByModelNameOrIdEvaluationByTaskIdParams,
+        params: GetAgentsBootstrapClassificationModelByJobIdParams,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -2471,31 +2553,47 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> EvaluationResponseClassificationMetrics:
+    ) -> BootstrapClassificationModelResponse:
+        """Get the status of a bootstrap classification model job"""
         pass
 
-    @overload
     def GET(
         self,
-        path: Literal["/regression_model"],
+        path: str,
         *,
-        params: None = None,
-        parse_as: Literal["json"] = "json",
+        params: Mapping[str, Any] | None = None,
+        parse_as: Literal["json", "text"] | None = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
         auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> list[RegressionModelMetadata]:
-        pass
+    ) -> Any:
+        path_params, query_params = self._parse_params(params or {}, path)
+        res = self.get(
+            path.format(**path_params),
+            params=query_params,
+            headers=headers,
+            cookies=cookies,
+            auth=auth,
+            follow_redirects=follow_redirects,
+            timeout=timeout,
+            extensions=extensions,
+        )
+        res.raise_for_status()
+        return (
+            None
+            if res.status_code == 204
+            else res.json() if parse_as == "json" else res.text if parse_as == "text" else res.content
+        )
 
     @overload
-    def GET(
+    def DELETE(
         self,
-        path: Literal["/regression_model/{name_or_id}"],
+        path: Literal["/cleanup"],
         *,
-        params: GetRegressionModelByNameOrIdParams,
+        params: None = None,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -2503,15 +2601,16 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> RegressionModelMetadata:
+    ) -> CleanupResponse:
+        """Cleanup orphaned milvus collections and blobs"""
         pass
 
     @overload
-    def GET(
+    def DELETE(
         self,
-        path: Literal["/regression_model/{model_name_or_id}/evaluation"],
+        path: Literal["/auth/api_key/{name_or_id}"],
         *,
-        params: GetRegressionModelByModelNameOrIdEvaluationParams,
+        params: DeleteAuthApiKeyByNameOrIdParams,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -2519,15 +2618,16 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> list[EvaluationResponseRegressionMetrics]:
+    ) -> None:
+        """Delete an API key by name or ID."""
         pass
 
     @overload
-    def GET(
+    def DELETE(
         self,
-        path: Literal["/regression_model/{model_name_or_id}/evaluation/{task_id}"],
+        path: Literal["/auth/org"],
         *,
-        params: GetRegressionModelByModelNameOrIdEvaluationByTaskIdParams,
+        params: None = None,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -2535,15 +2635,16 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> EvaluationResponseRegressionMetrics:
+    ) -> None:
+        """Deletes the org and all associated resources"""
         pass
 
     @overload
-    def GET(
+    def DELETE(
         self,
-        path: Literal["/task/{task_id}"],
+        path: Literal["/memoryset/{name_or_id}"],
         *,
-        params: GetTaskByTaskIdParams,
+        params: DeleteMemorysetByNameOrIdParams,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -2551,15 +2652,15 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> Task:
+    ) -> None:
         pass
 
     @overload
-    def GET(
+    def DELETE(
         self,
-        path: Literal["/task/{task_id}/status"],
+        path: Literal["/memoryset/{name_or_id}/memory/{memory_id}"],
         *,
-        params: GetTaskByTaskIdStatusParams,
+        params: DeleteMemorysetByNameOrIdMemoryByMemoryIdParams,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -2567,15 +2668,15 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> TaskStatusInfo:
+    ) -> None:
         pass
 
     @overload
-    def GET(
+    def DELETE(
         self,
-        path: Literal["/task"],
+        path: Literal["/finetuned_embedding_model/{name_or_id}"],
         *,
-        params: GetTaskParams,
+        params: DeleteFinetunedEmbeddingModelByNameOrIdParams,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -2583,15 +2684,16 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> PaginatedTask:
+    ) -> None:
+        """Delete a finetuned embedding model by name or ID."""
         pass
 
     @overload
-    def GET(
+    def DELETE(
         self,
-        path: Literal["/telemetry/prediction/{prediction_id}"],
+        path: Literal["/datasource/{name_or_id}"],
         *,
-        params: GetTelemetryPredictionByPredictionIdParams,
+        params: DeleteDatasourceByNameOrIdParams,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -2599,33 +2701,32 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> LabelPredictionWithMemoriesAndFeedback | ScorePredictionWithMemoriesAndFeedback:
-        """Get a specific prediction by ID."""
+    ) -> None:
+        """Delete a datasource by name or ID."""
         pass
 
     @overload
-    def GET(
+    def DELETE(
         self,
-        path: Literal["/telemetry/prediction/{prediction_id}/explanation"],
+        path: Literal["/classification_model/{name_or_id}"],
         *,
-        params: GetTelemetryPredictionByPredictionIdExplanationParams,
-        parse_as: Literal["text"],
+        params: DeleteClassificationModelByNameOrIdParams,
+        parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
         auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> str:
-        """Get explanation for a prediction, optionally streaming the response."""
+    ) -> None:
         pass
 
     @overload
-    def GET(
+    def DELETE(
         self,
-        path: Literal["/telemetry/prediction/{prediction_id}/action"],
+        path: Literal["/regression_model/{name_or_id}"],
         *,
-        params: GetTelemetryPredictionByPredictionIdActionParams,
+        params: DeleteRegressionModelByNameOrIdParams,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -2633,16 +2734,15 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> ActionRecommendation:
-        """Get action recommendation for improving a specific prediction."""
+    ) -> None:
         pass
 
     @overload
-    def GET(
+    def DELETE(
         self,
-        path: Literal["/telemetry/prediction/{prediction_id}/memory_suggestions"],
+        path: Literal["/classification_model/{model_name_or_id}/evaluation/{job_id}"],
         *,
-        params: GetTelemetryPredictionByPredictionIdMemorySuggestionsParams,
+        params: DeleteClassificationModelByModelNameOrIdEvaluationByJobIdParams,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -2650,21 +2750,15 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> AddMemoryRecommendations:
-        """
-        Generate synthetic memory suggestions to improve a specific prediction.
-
-        The returned suggestions have labels as string representations of integer indices
-        corresponding to the memoryset's label_names.
-        """
+    ) -> None:
         pass
 
     @overload
-    def GET(
+    def DELETE(
         self,
-        path: Literal["/telemetry/feedback_category"],
+        path: Literal["/regression_model/{model_name_or_id}/evaluation/{job_id}"],
         *,
-        params: None = None,
+        params: DeleteRegressionModelByModelNameOrIdEvaluationByJobIdParams,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -2672,16 +2766,15 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> list[PredictionFeedbackCategory]:
-        """List all feedback categories for the organization."""
+    ) -> None:
         pass
 
     @overload
-    def GET(
+    def DELETE(
         self,
-        path: Literal["/telemetry/feedback_category/{name_or_id}"],
+        path: Literal["/job/{job_id}/abort"],
         *,
-        params: GetTelemetryFeedbackCategoryByNameOrIdParams,
+        params: DeleteJobByJobIdAbortParams,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -2689,16 +2782,15 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> PredictionFeedbackCategory:
-        """Get a feedback category by name or ID."""
+    ) -> None:
         pass
 
     @overload
-    def GET(
+    def DELETE(
         self,
-        path: Literal["/agents/bootstrap_classification_model/{task_id}"],
+        path: Literal["/telemetry/feedback_category/{name_or_id}"],
         *,
-        params: GetAgentsBootstrapClassificationModelByTaskIdParams,
+        params: DeleteTelemetryFeedbackCategoryByNameOrIdParams,
         parse_as: Literal["json"] = "json",
         headers: HeaderTypes | None = None,
         cookies: CookieTypes | None = None,
@@ -2706,11 +2798,11 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> BootstrapClassificationModelResponse:
-        """Get the status of a bootstrap classification model task"""
+    ) -> None:
+        """Delete a feedback category and all associated feedback records."""
         pass
 
-    def GET(
+    def DELETE(
         self,
         path: str,
         *,
@@ -2724,7 +2816,7 @@ class OrcaClient(Client):
         extensions: RequestExtensions | None = None,
     ) -> Any:
         path_params, query_params = self._parse_params(params or {}, path)
-        res = self.get(
+        res = self.delete(
             path.format(**path_params),
             params=query_params,
             headers=headers,
@@ -2733,7 +2825,8 @@ class OrcaClient(Client):
             follow_redirects=follow_redirects,
             timeout=timeout,
             extensions=extensions,
-        ).raise_for_status()
+        )
+        res.raise_for_status()
         return (
             None
             if res.status_code == 204
@@ -2842,6 +2935,26 @@ class OrcaClient(Client):
     ) -> None:
         pass
 
+    @overload
+    def POST(
+        self,
+        path: Literal["/gpu/memoryset/{name_or_id}/lookup"],
+        *,
+        params: PostGpuMemorysetByNameOrIdLookupParams,
+        json: LookupRequest,
+        data: None = None,
+        files: None = None,
+        content: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> list[list[LabeledMemoryLookup | ScoredMemoryLookup]]:
+        pass
+
     @overload
     def POST(
         self,
@@ -2859,16 +2972,76 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> list[LabeledMemory] | list[ScoredMemory]:
+    ) -> list[LabeledMemory] | list[ScoredMemory]:
+        pass
+
+    @overload
+    def POST(
+        self,
+        path: Literal["/memoryset/{name_or_id}/memories"],
+        *,
+        params: PostMemorysetByNameOrIdMemoriesParams,
+        json: ListMemoriesRequest | None = None,
+        data: None = None,
+        files: None = None,
+        content: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> list[LabeledMemory] | list[ScoredMemory]:
+        pass
+
+    @overload
+    def POST(
+        self,
+        path: Literal["/memoryset/{name_or_id}/memories/delete"],
+        *,
+        params: PostMemorysetByNameOrIdMemoriesDeleteParams,
+        json: DeleteMemoriesRequest,
+        data: None = None,
+        files: None = None,
+        content: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> None:
+        pass
+
+    @overload
+    def POST(
+        self,
+        path: Literal["/gpu/memoryset/{name_or_id}/memory"],
+        *,
+        params: PostGpuMemorysetByNameOrIdMemoryParams,
+        json: PostGpuMemorysetByNameOrIdMemoryRequest,
+        data: None = None,
+        files: None = None,
+        content: None = None,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> list[str]:
         pass
 
     @overload
     def POST(
         self,
-        path: Literal["/memoryset/{name_or_id}/memories"],
+        path: Literal["/memoryset/{name_or_id}/analysis"],
         *,
-        params: PostMemorysetByNameOrIdMemoriesParams,
-        json: ListMemoriesRequest | None = None,
+        params: PostMemorysetByNameOrIdAnalysisParams,
+        json: MemorysetAnalysisRequest,
         data: None = None,
         files: None = None,
         content: None = None,
@@ -2879,16 +3052,16 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> list[LabeledMemory] | list[ScoredMemory]:
+    ) -> MemorysetAnalysisResponse:
         pass
 
     @overload
     def POST(
         self,
-        path: Literal["/memoryset/{name_or_id}/memories/delete"],
+        path: Literal["/memoryset/{name_or_id}/memory/{memory_id}/cascading_edits"],
         *,
-        params: PostMemorysetByNameOrIdMemoriesDeleteParams,
-        json: DeleteMemoriesRequest,
+        params: PostMemorysetByNameOrIdMemoryByMemoryIdCascadingEditsParams,
+        json: CascadeEditSuggestionsRequest,
         data: None = None,
         files: None = None,
         content: None = None,
@@ -2899,16 +3072,16 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> None:
+    ) -> list[CascadingEditSuggestion]:
         pass
 
     @overload
     def POST(
         self,
-        path: Literal["/memoryset/{name_or_id}/analysis"],
+        path: Literal["/finetuned_embedding_model"],
         *,
-        params: PostMemorysetByNameOrIdAnalysisParams,
-        json: MemorysetAnalysisRequest,
+        params: None = None,
+        json: FinetuneEmbeddingModelRequest,
         data: None = None,
         files: None = None,
         content: None = None,
@@ -2919,16 +3092,17 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> MemorysetAnalysisResponse:
+    ) -> FinetunedEmbeddingModelMetadata:
+        """Create a finetuned embedding model."""
         pass
 
     @overload
     def POST(
         self,
-        path: Literal["/memoryset/{name_or_id}/memory/{memory_id}/cascading_edits"],
+        path: Literal["/gpu/finetuned_embedding_model/{name_or_id}/embedding"],
         *,
-        params: PostMemorysetByNameOrIdMemoryByMemoryIdCascadingEditsParams,
-        json: CascadeEditSuggestionsRequest,
+        params: PostGpuFinetunedEmbeddingModelByNameOrIdEmbeddingParams,
+        json: EmbedRequest,
         data: None = None,
         files: None = None,
         content: None = None,
@@ -2939,16 +3113,17 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> list[CascadingEditSuggestion]:
+    ) -> list[list[float]]:
+        """Embed values using a finetuned embedding model."""
         pass
 
     @overload
     def POST(
         self,
-        path: Literal["/finetuned_embedding_model"],
+        path: Literal["/gpu/pretrained_embedding_model/{model_name}/embedding"],
         *,
-        params: None = None,
-        json: FinetuneEmbeddingModelRequest,
+        params: PostGpuPretrainedEmbeddingModelByModelNameEmbeddingParams,
+        json: EmbedRequest,
         data: None = None,
         files: None = None,
         content: None = None,
@@ -2959,8 +3134,8 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> FinetunedEmbeddingModelMetadata:
-        """Create a finetuned embedding model."""
+    ) -> list[list[float]]:
+        """Embed values using a pretrained embedding model."""
         pass
 
     @overload
@@ -3062,10 +3237,10 @@ class OrcaClient(Client):
     @overload
     def POST(
         self,
-        path: Literal["/datasource/{name_or_id}/embedding_evaluation"],
+        path: Literal["/datasource/{name_or_id}/rows"],
         *,
-        params: PostDatasourceByNameOrIdEmbeddingEvaluationParams,
-        json: DatasourceEmbeddingEvaluationsRequest,
+        params: PostDatasourceByNameOrIdRowsParams,
+        json: GetDatasourceRowsRequest,
         data: None = None,
         files: None = None,
         content: None = None,
@@ -3076,17 +3251,17 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> DatasourceEmbeddingEvaluationsResponse:
-        """Create an embedding evaluation task for a datasource."""
+    ) -> list[dict[str, Any]]:
+        """Get rows from a specific datasource with optional filtering."""
         pass
 
     @overload
     def POST(
         self,
-        path: Literal["/classification_model"],
+        path: Literal["/datasource/{name_or_id}/rows/count"],
         *,
-        params: None = None,
-        json: CreateClassificationModelRequest,
+        params: PostDatasourceByNameOrIdRowsCountParams,
+        json: GetDatasourceRowCountRequest,
         data: None = None,
         files: None = None,
         content: None = None,
@@ -3097,16 +3272,17 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> ClassificationModelMetadata:
+    ) -> int:
+        """Get row count from a specific datasource with optional filtering."""
         pass
 
     @overload
     def POST(
         self,
-        path: Literal["/classification_model/{model_name_or_id}/evaluation"],
+        path: Literal["/classification_model"],
         *,
-        params: PostClassificationModelByModelNameOrIdEvaluationParams,
-        json: ClassificationEvaluationRequest,
+        params: None = None,
+        json: CreateClassificationModelRequest,
         data: None = None,
         files: None = None,
         content: None = None,
@@ -3117,7 +3293,7 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> EvaluationResponse:
+    ) -> ClassificationModelMetadata:
         pass
 
     @overload
@@ -3143,10 +3319,10 @@ class OrcaClient(Client):
     @overload
     def POST(
         self,
-        path: Literal["/regression_model/{model_name_or_id}/evaluation"],
+        path: Literal["/gpu/classification_model/{name_or_id}/prediction"],
         *,
-        params: PostRegressionModelByModelNameOrIdEvaluationParams,
-        json: RegressionEvaluationRequest,
+        params: PostGpuClassificationModelByNameOrIdPredictionParams,
+        json: ClassificationPredictionRequest,
         data: None = None,
         files: None = None,
         content: None = None,
@@ -3157,16 +3333,16 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> EvaluationResponse:
+    ) -> list[BaseLabelPredictionResult]:
         pass
 
     @overload
     def POST(
         self,
-        path: Literal["/telemetry/prediction"],
+        path: Literal["/classification_model/{name_or_id}/prediction"],
         *,
-        params: None = None,
-        json: ListPredictionsRequest | None = None,
+        params: PostClassificationModelByNameOrIdPredictionParams,
+        json: ClassificationPredictionRequest,
         data: None = None,
         files: None = None,
         content: None = None,
@@ -3177,17 +3353,16 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> list[LabelPredictionWithMemoriesAndFeedback | ScorePredictionWithMemoriesAndFeedback]:
-        """List predictions with optional filtering and sorting."""
+    ) -> list[BaseLabelPredictionResult]:
         pass
 
     @overload
     def POST(
         self,
-        path: Literal["/telemetry/prediction/count"],
+        path: Literal["/gpu/regression_model/{name_or_id}/prediction"],
         *,
-        params: None = None,
-        json: CountPredictionsRequest | None = None,
+        params: PostGpuRegressionModelByNameOrIdPredictionParams,
+        json: RegressionPredictionRequest,
         data: None = None,
         files: None = None,
         content: None = None,
@@ -3198,17 +3373,16 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> int:
-        """Count predictions with optional filtering."""
+    ) -> list[BaseScorePredictionResult]:
         pass
 
     @overload
     def POST(
         self,
-        path: Literal["/telemetry/memories"],
+        path: Literal["/regression_model/{name_or_id}/prediction"],
         *,
-        params: None = None,
-        json: TelemetryMemoriesRequest,
+        params: PostRegressionModelByNameOrIdPredictionParams,
+        json: RegressionPredictionRequest,
         data: None = None,
         files: None = None,
         content: None = None,
@@ -3219,21 +3393,16 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> PaginatedUnionLabeledMemoryWithFeedbackMetricsScoredMemoryWithFeedbackMetrics:
-        """
-        List memories with feedback metrics.
-        **Note**: This endpoint will ONLY return memories that have been used in a prediction.
-        If you want to query ALL memories WITHOUT feedback metrics, use the query_memoryset endpoint.
-        """
+    ) -> list[BaseScorePredictionResult]:
         pass
 
     @overload
     def POST(
         self,
-        path: Literal["/agents/bootstrap_classification_model"],
+        path: Literal["/classification_model/{model_name_or_id}/evaluation"],
         *,
-        params: None = None,
-        json: BootstrapClassificationModelRequest,
+        params: PostClassificationModelByModelNameOrIdEvaluationParams,
+        json: ClassificationEvaluationRequest,
         data: None = None,
         files: None = None,
         content: None = None,
@@ -3244,30 +3413,16 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> BootstrapClassificationModelResponse:
-        """
-        Bootstrap a classification model by creating a memoryset with generated memories and a classification model.
-
-        This endpoint uses the bootstrap_classification_model agent to generate:
-        1. Memoryset configuration with appropriate settings
-        2. Model configuration with optimal parameters
-        3. High-quality training memories for each label
-
-        The process involves:
-        1. Calling the agent to generate configurations and memories
-        2. Creating a datasource from the generated memories
-        3. Creating a memoryset from the datasource
-        4. Creating a classification model from the memoryset
-        """
+    ) -> EvaluationResponse:
         pass
 
     @overload
     def POST(
         self,
-        path: Literal["/gpu/memoryset/{name_or_id}/lookup"],
+        path: Literal["/regression_model/{model_name_or_id}/evaluation"],
         *,
-        params: PostGpuMemorysetByNameOrIdLookupParams,
-        json: LookupRequest,
+        params: PostRegressionModelByModelNameOrIdEvaluationParams,
+        json: RegressionEvaluationRequest,
         data: None = None,
         files: None = None,
         content: None = None,
@@ -3278,16 +3433,16 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> list[list[LabeledMemoryLookup | ScoredMemoryLookup]]:
+    ) -> EvaluationResponse:
         pass
 
     @overload
     def POST(
         self,
-        path: Literal["/gpu/memoryset/{name_or_id}/memory"],
+        path: Literal["/telemetry/prediction"],
         *,
-        params: PostGpuMemorysetByNameOrIdMemoryParams,
-        json: PostGpuMemorysetByNameOrIdMemoryRequest,
+        params: None = None,
+        json: ListPredictionsRequest | None = None,
         data: None = None,
         files: None = None,
         content: None = None,
@@ -3298,16 +3453,17 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> list[str]:
+    ) -> list[LabelPredictionWithMemoriesAndFeedback | ScorePredictionWithMemoriesAndFeedback]:
+        """List predictions with optional filtering and sorting."""
         pass
 
     @overload
     def POST(
         self,
-        path: Literal["/gpu/classification_model/{name_or_id}/prediction"],
+        path: Literal["/telemetry/prediction/count"],
         *,
-        params: PostGpuClassificationModelByNameOrIdPredictionParams,
-        json: ClassificationPredictionRequest,
+        params: None = None,
+        json: CountPredictionsRequest | None = None,
         data: None = None,
         files: None = None,
         content: None = None,
@@ -3318,16 +3474,17 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> list[BaseLabelPredictionResult]:
+    ) -> int:
+        """Count predictions with optional filtering."""
         pass
 
     @overload
     def POST(
         self,
-        path: Literal["/gpu/regression_model/{name_or_id}/prediction"],
+        path: Literal["/telemetry/memories"],
         *,
-        params: PostGpuRegressionModelByNameOrIdPredictionParams,
-        json: RegressionPredictionRequest,
+        params: None = None,
+        json: TelemetryMemoriesRequest,
         data: None = None,
         files: None = None,
         content: None = None,
@@ -3338,16 +3495,21 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> list[BaseScorePredictionResult]:
+    ) -> PaginatedUnionLabeledMemoryWithFeedbackMetricsScoredMemoryWithFeedbackMetrics:
+        """
+        List memories with feedback metrics.
+        **Note**: This endpoint will ONLY return memories that have been used in a prediction.
+        If you want to query ALL memories WITHOUT feedback metrics, use the query_memoryset endpoint.
+        """
         pass
 
     @overload
     def POST(
         self,
-        path: Literal["/gpu/finetuned_embedding_model/{name_or_id}/embedding"],
+        path: Literal["/agents/bootstrap_classification_model"],
         *,
-        params: PostGpuFinetunedEmbeddingModelByNameOrIdEmbeddingParams,
-        json: EmbedRequest,
+        params: None = None,
+        json: BootstrapClassificationModelRequest,
         data: None = None,
         files: None = None,
         content: None = None,
@@ -3358,29 +3520,21 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> list[list[float]]:
-        """Embed values using a finetuned embedding model."""
-        pass
+    ) -> BootstrapClassificationModelResponse:
+        """
+        Bootstrap a classification model by creating a memoryset with generated memories and a classification model.
 
-    @overload
-    def POST(
-        self,
-        path: Literal["/gpu/pretrained_embedding_model/{model_name}/embedding"],
-        *,
-        params: PostGpuPretrainedEmbeddingModelByModelNameEmbeddingParams,
-        json: EmbedRequest,
-        data: None = None,
-        files: None = None,
-        content: None = None,
-        parse_as: Literal["json"] = "json",
-        headers: HeaderTypes | None = None,
-        cookies: CookieTypes | None = None,
-        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
-        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
-        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
-        extensions: RequestExtensions | None = None,
-    ) -> list[list[float]]:
-        """Embed values using a pretrained embedding model."""
+        This endpoint uses the bootstrap_classification_model agent to generate:
+        1. Memoryset configuration with appropriate settings
+        2. Model configuration with optimal parameters
+        3. High-quality training memories for each label
+
+        The process involves:
+        1. Calling the agent to generate configurations and memories
+        2. Creating a datasource from the generated memories
+        3. Creating a memoryset from the datasource
+        4. Creating a classification model from the memoryset
+        """
         pass
 
     def POST(
@@ -3414,7 +3568,8 @@ class OrcaClient(Client):
             follow_redirects=follow_redirects,
             timeout=timeout,
             extensions=extensions,
-        ).raise_for_status()
+        )
+        res.raise_for_status()
         return (
             None
             if res.status_code == 204
@@ -3494,7 +3649,8 @@ class OrcaClient(Client):
             follow_redirects=follow_redirects,
             timeout=timeout,
             extensions=extensions,
-        ).raise_for_status()
+        )
+        res.raise_for_status()
         return (
             None
             if res.status_code == 204
@@ -3524,10 +3680,10 @@ class OrcaClient(Client):
     @overload
     def PATCH(
         self,
-        path: Literal["/classification_model/{name_or_id}"],
+        path: Literal["/gpu/memoryset/{name_or_id}/memory"],
         *,
-        params: PatchClassificationModelByNameOrIdParams,
-        json: PredictiveModelUpdate,
+        params: PatchGpuMemorysetByNameOrIdMemoryParams,
+        json: PatchGpuMemorysetByNameOrIdMemoryRequest,
         data: None = None,
         files: None = None,
         content: None = None,
@@ -3538,16 +3694,16 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> ClassificationModelMetadata:
+    ) -> LabeledMemory | ScoredMemory:
         pass
 
     @overload
     def PATCH(
         self,
-        path: Literal["/regression_model/{name_or_id}"],
+        path: Literal["/gpu/memoryset/{name_or_id}/memories"],
         *,
-        params: PatchRegressionModelByNameOrIdParams,
-        json: PredictiveModelUpdate,
+        params: PatchGpuMemorysetByNameOrIdMemoriesParams,
+        json: PatchGpuMemorysetByNameOrIdMemoriesRequest,
         data: None = None,
         files: None = None,
         content: None = None,
@@ -3558,16 +3714,16 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> RegressionModelMetadata:
+    ) -> list[LabeledMemory] | list[ScoredMemory]:
         pass
 
     @overload
     def PATCH(
         self,
-        path: Literal["/telemetry/prediction/{prediction_id}"],
+        path: Literal["/classification_model/{name_or_id}"],
         *,
-        params: PatchTelemetryPredictionByPredictionIdParams,
-        json: UpdatePredictionRequest,
+        params: PatchClassificationModelByNameOrIdParams,
+        json: PredictiveModelUpdate,
         data: None = None,
         files: None = None,
         content: None = None,
@@ -3578,17 +3734,16 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> Any:
-        """Update a prediction with new expected values, tags, or memory ID."""
+    ) -> ClassificationModelMetadata:
         pass
 
     @overload
     def PATCH(
         self,
-        path: Literal["/gpu/memoryset/{name_or_id}/memory"],
+        path: Literal["/regression_model/{name_or_id}"],
         *,
-        params: PatchGpuMemorysetByNameOrIdMemoryParams,
-        json: PatchGpuMemorysetByNameOrIdMemoryRequest,
+        params: PatchRegressionModelByNameOrIdParams,
+        json: PredictiveModelUpdate,
         data: None = None,
         files: None = None,
         content: None = None,
@@ -3599,16 +3754,16 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> LabeledMemory | ScoredMemory:
+    ) -> RegressionModelMetadata:
         pass
 
     @overload
     def PATCH(
         self,
-        path: Literal["/gpu/memoryset/{name_or_id}/memories"],
+        path: Literal["/telemetry/prediction/{prediction_id}"],
         *,
-        params: PatchGpuMemorysetByNameOrIdMemoriesParams,
-        json: PatchGpuMemorysetByNameOrIdMemoriesRequest,
+        params: PatchTelemetryPredictionByPredictionIdParams,
+        json: UpdatePredictionRequest,
         data: None = None,
         files: None = None,
         content: None = None,
@@ -3619,7 +3774,8 @@ class OrcaClient(Client):
         follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
-    ) -> list[LabeledMemory] | list[ScoredMemory]:
+    ) -> Any:
+        """Update a prediction with new expected values, tags, or memory ID."""
         pass
 
     def PATCH(
@@ -3653,95 +3809,126 @@ class OrcaClient(Client):
             follow_redirects=follow_redirects,
             timeout=timeout,
             extensions=extensions,
-        ).raise_for_status()
+        )
+        res.raise_for_status()
         return (
             None
             if res.status_code == 204
             else res.json() if parse_as == "json" else res.text if parse_as == "text" else res.content
         )
 
-    @property
-    def base_url(self) -> URL:
-        # lazy init base_url from environment variable
-        if str(self._base_url) == "":
-            self.base_url = os.environ.get("ORCA_API_URL", "https://api.orcadb.ai/")
-        return self._base_url
+    @staticmethod
+    def _raise_error_for_response(response: Response) -> None:
+        if response.status_code == 401:
+            raise ValueError("Invalid API key")
+        # elif response.status_code == 402:
+        #     res = cast(QuotaExceededErrorResponse, json.loads(response.read().decode(response.encoding or "utf-8")))
+        #     raise RuntimeError(
+        #         f"{res['quota_type'].replace('_', ' ').title()} limit reached ({res['current']}/{res['quota_limit']})"
+        #     )
+        elif response.status_code == 403:
+            raise PermissionError(json.loads(response.read().decode(response.encoding or "utf-8"))["reason"])
+        elif response.status_code == 404:
+            res = cast(NotFoundErrorResponse, json.loads(response.read().decode(response.encoding or "utf-8")))
+            if res["resource"] is not None:
+                raise LookupError(f"The {res['resource']} you are looking for does not exist")
+            else:
+                raise RuntimeError(f"Unknown API route: {response.url}")
+        elif response.status_code == 405:
+            raise RuntimeError(f"Unknown method {response.request.method} for API route: {response.url}")
+        elif response.status_code == 409:
+            res = cast(
+                ConstraintViolationErrorResponse, json.loads(response.read().decode(response.encoding or "utf-8"))
+            )
+            raise RuntimeError(res["constraint"])
+        elif response.status_code == 422:
+            res = cast(InvalidInputErrorResponse, json.loads(response.read().decode(response.encoding or "utf-8")))
+            issues = [f"{issue['loc'][-1]}: {issue['msg']}" for issue in res["validation_issues"]]
+            raise ValueError("Invalid input:\n\t" + "\n\t".join(issues))
+        elif response.status_code == 500:
+            res = cast(InternalServerErrorResponse, json.loads(response.read().decode(response.encoding or "utf-8")))
+            raise RuntimeError(f"Unexpected server error: {res['message']}")
+        elif response.status_code == 503:
+            raise RuntimeError("Orca API is currently unavailable, please try again later")
+        elif response.status_code >= 400:
+            raise RuntimeError(f"Unexpected status code: {response.status_code}")
+
+    @staticmethod
+    def _instrument_request(request: Request) -> None:
+        request.headers["X-Request-ID"] = str(uuid.uuid4())
+
+    def __init__(
+        self,
+        *,
+        api_key: str | None = None,
+        base_url: URL | str = "",
+        headers: HeaderTypes | None = None,
+        transport: BaseTransport | None = None,
+        timeout: TimeoutTypes | None = None,
+        limits: Limits | None = None,
+        max_redirects: int = 20,
+        event_hooks: None | (Mapping[str, list[Callable[..., Any]]]) = None,
+        http1: bool = True,
+        http2: bool = False,
+        proxy: str | URL | Proxy | None = None,
+        log_level: int = logging.WARNING,
+    ) -> None:
+        """
+        Initialize an OrcaAPI httpx client
 
-    @base_url.setter
-    def base_url(self, url: URL | str) -> None:
-        Client.base_url.__set__(self, url)
+        Params:
+            api_key: API key to use for authentication, will default to ORCA_API_KEY if not set.
+            base_url: URL of the OrcaAPI, will default to ORCA_API_URL or the cloud API URL if not set.
+        """
+        logging.getLogger("httpx").setLevel(log_level)
+        logging.getLogger("httpcore").setLevel(log_level)
+        super().__init__(
+            headers={"Api-Key": api_key or os.environ.get("ORCA_API_KEY", "")} | dict(Headers(headers or {}).items()),
+            http1=http1,
+            http2=http2,
+            proxy=proxy,
+            timeout=timeout or Timeout(connect=3, read=20, write=10, pool=5),
+            follow_redirects=True,
+            limits=limits or Limits(max_connections=100, max_keepalive_connections=20),
+            max_redirects=max_redirects,
+            event_hooks=event_hooks
+            or {"request": [self._instrument_request], "response": [self._raise_error_for_response]},
+            base_url=base_url or os.environ.get("ORCA_API_URL", "https://api.orcadb.ai/"),
+            transport=transport
+            or RetryTransport(
+                transport=HTTPTransport(),
+                retry=Retry(
+                    total=5,
+                    backoff_factor=0.5,
+                    allowed_methods=["GET", "POST", "PUT", "PATCH", "DELETE"],
+                    status_forcelist=[429, 500, 502, 503, 504],
+                ),
+            ),
+        )
 
     @property
-    def headers(self) -> Headers:
-        # lazy init headers from environment variable
-        if not self._headers.get("Api-Key"):
-            Client.headers.__set__(self, Headers({"Api-Key": os.environ.get("ORCA_API_KEY", "")}))
-        return self._headers
-
-    @headers.setter
-    def headers(self, headers: HeaderTypes) -> None:
-        Client.headers.__set__(self, headers)
-
-
-def _read_json_response(response: Response) -> dict[str, Any]:
-    content = response.read()
-    text = content.decode(response.encoding or "utf-8")
-    return json.loads(text)
-
-
-def _raise_error_for_response(response: Response) -> None:
-    if response.status_code == 401:
-        raise ValueError("Invalid API key")
-    # elif response.status_code == 402:
-    #     res = cast(QuotaExceededErrorResponse, _read_json_response(response))
-    #     raise RuntimeError(
-    #         f"{res['quota_type'].replace('_', ' ').title()} limit reached ({res['current']}/{res['quota_limit']})"
-    #     )
-    elif response.status_code == 403:
-        raise PermissionError(_read_json_response(response)["reason"])
-    elif response.status_code == 404:
-        res = cast(NotFoundErrorResponse, _read_json_response(response))
-        if res["resource"] is not None:
-            raise LookupError(f"The {res['resource']} you are looking for does not exist")
-        else:
-            raise RuntimeError(f"Unknown API route: {response.url}")
-    elif response.status_code == 405:
-        raise RuntimeError(f"Unknown method {response.request.method} for API route: {response.url}")
-    elif response.status_code == 409:
-        res = cast(ConstraintViolationErrorResponse, _read_json_response(response))
-        raise RuntimeError(res["constraint"])
-    elif response.status_code == 422:
-        res = cast(InvalidInputErrorResponse, _read_json_response(response))
-        issues = [f"{issue['loc'][-1]}: {issue['msg']}" for issue in res["validation_issues"]]
-        raise ValueError("Invalid input:\n\t" + "\n\t".join(issues))
-    elif response.status_code == 500:
-        res = cast(InternalServerErrorResponse, _read_json_response(response))
-        raise RuntimeError(f"Unexpected server error: {res['message']}")
-    elif response.status_code == 503:
-        raise RuntimeError("Orca API is currently unavailable, please try again later")
-    elif response.status_code >= 400:
-        raise RuntimeError(f"Unexpected status code: {response.status_code}")
-
-
-def _instrument_request(request: Request) -> None:
-    request.headers["X-Request-ID"] = str(uuid.uuid4())
-
-
-logging.getLogger("httpx").setLevel(logging.WARNING)
-logging.getLogger("httpcore").setLevel(logging.ERROR)
-
-orca_api = OrcaClient(
-    transport=RetryTransport(
-        transport=HTTPTransport(),
-        retry=Retry(
-            total=5,
-            backoff_factor=0.5,
-            allowed_methods=["GET", "POST", "PUT", "PATCH", "DELETE"],
-            status_forcelist=[429, 500, 502, 503, 504],
-        ),
-    ),
-    event_hooks={"request": [_instrument_request], "response": [_raise_error_for_response]},
-    follow_redirects=True,
-    timeout=Timeout(connect=3, read=20, write=10, pool=5),
-)
-"""Typed client for the Orca API"""
+    def api_key(self) -> str:
+        return self.headers["Api-Key"]
+
+    @api_key.setter
+    def api_key(self, api_key: str) -> None:
+        self.headers.update(Headers({"Api-Key": api_key}))
+
+    client_ctx = ContextVar[Self | None]("orca_client", default=None)
+    default_client: Self | None = None
+
+    @contextmanager
+    def use(self) -> Generator[None, None, None]:
+        """Context manager to inject this client into any OrcaSDK methods"""
+        token = self.client_ctx.set(self)
+        try:
+            yield
+        finally:
+            self.client_ctx.reset(token)
+
+    @classmethod
+    def _resolve_client(cls, client: Self | None = None) -> Self:
+        client = client or cls.client_ctx.get() or cls.default_client
+        if not client:
+            client = cls.default_client = cls()
+        return client